Curve and surface thinning are widely-used skeletonization techniques for modeling objects in 3 dimensions. In the case of disordered porous media analysis, however, neither is really efficient since the internal geometry of the object is usually composed of both rod and plate shapes. This paper concludes an application of discrete wavelet transform (WT) and complex wavelet transform (CWT) in image processing problem such as hybrid skeletonization of trabecular bone images. Hybrid skeleton combines 2D surfaces and 1D curve to represent respectively the plate-shaped and rod-shaped parts of the object. For hybrid skeletonization, two cascade structures are proposed. In these structures, features of images were extracted with discrete wavelet transform and complex wavelet transform. After that, obtained features were used as inputs of complex-valued artificial neural network (CVANN) which is multilayered artificial neural networks with two dimensions (real and imaginary parts). Effects of the feature extraction methods are compared for ability of the hybrid skeletonization on a trabecular bone sample. Results show that the CWT succeeded to hybrid skeletonization with lower error rate than WT.